Human-driven vehicles' cruising versus autonomous vehicles' backand-forth congestion: The effects on traveling, parking and congestion

This paper explores how the interaction between human-driven vehicles (HVs) cruising for parking and autonomous vehicles (AVs) traveling back and forth affects travel behavior and congestion. To capture the spatial distribution of parking, we develop a continuous spatial optimization model, with a discrete choice logit model governing the choice between the two modes. Various congestion externalities are considered in the proposed model. Using optimal control method, we derive the social optimum under user-equilibrium constraints and compare it to the unpriced user equilibrium. Without pricing, the introduction of AVs may increase or lower congestion depending on whether cruising or traveling back and forth dominates. Thus, AVs may be underused or overused, as the marginal external benefit of switching to AVs may be positive or negative. In our numerical model, with optimal pricing, the introduction of AVs always reduces travel costs. In terms of parking, the introduction of AVs is efficient in reducing parking demand and results in a smaller and less compact city. The efficiency of pricing is significantly impacted by the congestion interactions. When only one congestion type exists, it increases with the degree of congestion. However, when both congestion types exist, the efficiency of pricing is ambiguous. Specifically, when both HVs' cruising congestion and AVs' back-and-forth congestion are heavy, the efficiency of pricing is lower. Our proposed model reveals that the effects of AVs on travel and urban equilibrium may be more difficult to assess and less beneficial than often thought. Our numerical study provides policy insights into actions that regulators could consider for the operation of AVs.